JP7215601B2 - Processing device, relocation method and relocation program - Google Patents

Description

The present invention relates to a processing device, a rearrangement method, and a rearrangement program.

Stateful network functions such as NAPT (Network Address Port Translation) and firewalls that use state information, which is past processing history, are widely used for packet processing. Now, by implementing these packet processing functions in software and running them on a general-purpose server, it is now possible to deploy functions according to traffic demand. Furthermore, it has become possible to optimize the transfer route associated with the movement of the user's terminal, which is the delivery destination of the data, by moving (relocating) the software implementation to another server.

Among such software, when relocating software that executes a stateful packet processing function, it is necessary to transfer data to the relocation destination for resuming the software at the relocation destination. The data transferred to the relocation destination is the flow state and the like.

This flow state is updated by packet processing. As a result, the flow state transferred to the relocation destination while packet processing continues is continuously updated by the software of the relocation source at the relocation source. It's getting old. The software needs to process packets based on the latest state, but there was a problem that the software could not be operated correctly at the relocation destination after the relocation of the packet processing functions.

Therefore, conventionally, it is necessary to suspend the software in order to completely synchronize the flow states of the relocation source and the relocation destination. However, packets arriving during software suspension are dropped or wait in a queue, resulting in a problem of degraded transfer quality.

In particular, in services such as VR (Virtual Reality) and AR (Augmented Reality), events that occur around the user are reflected in the virtual space in real time, so the transfer time for event-related data and virtual space images is shortened. There is a need. However, when network functions are rearranged when services such as VR and AR are provided, a phenomenon occurs that causes an increase in transfer time due to the aforementioned data drop, queuing delay, and the like.

13 and 14 are diagrams for explaining conventional data transfer processing. As an effective data transfer delay reduction method when the packet processing function is implemented in software, multiple packet processing functions that process the same data flow are operated on the same server, and the packets transferred between functions using the memory bus are reduced. There is a method for high-speed transfer (see (1) in FIG. 13). When such a packet processing function is moved to another server, the data transferred to the packet processing function whose movement completion timing is earlier will be routed through the network, which has a longer transfer delay than the memory bus (Fig. 13 (2)).

Also, as shown in FIG. 14, if the traffic is processed in a predetermined order, each time a packet being transferred arrives at the software performing the relocation, it will suffer a queuing delay. Specifically, as shown in FIG. 14, the packets queued in the upstream packet processing function are transferred to the downstream packet processing function after the flow state transition of the upstream packet processing function is completed. However, if the flow state is changed also in the downstream packet processing function, queuing is also performed downstream, resulting in double queuing delay. Therefore, it is required to suppress an increase in transfer delay caused by these factors when providing services that require low-delay communication.

Therefore, conventionally, there has been proposed a method for shortening the flow transfer stop time when network functions are rearranged when a plurality of network functions process the same flow (see, for example, Non-Patent Document 1).

K. Sugisono, et al., “Migration for VNF instances Forming Service Chain,” IEEE CloudNet2018, 2018.

The method described in Non-Patent Document 1 is a case in which the pause time of a packet processing function at the time of relocation greatly depends on the amount of packets processed by the packet processing function, and packet queuing is performed by each packet processing function. It is a method for shortening the time in which packet loss occurs if not performed. In the method described in Non-Patent Document 1, a scheduler determines the packet processing function to be rearranged and the timing of execution of the rearrangement.

Therefore, the method described in Non-Patent Document 1 was effective when packets were dropped during rearrangement and the increase in transfer delay as described above did not occur. However, when the order of traffic flow to the packet processing functions is determined, the packets are transferred to the destination by queuing the packets during the rearrangement of the packet processing functions. could not be applied and the transfer delay that occurred during that time could not be reduced.

Therefore, the method described in Non-Patent Document 1 has a problem that packet queuing delay and delay increase due to packets passing through the network cannot be avoided.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a processing device, a relocation method, and a relocation program capable of reducing transfer delay that occurs during relocation of data processing functions. .

In order to solve the above-described problems and achieve the object, the processing apparatus of the present invention is a processing apparatus that relocates the data processing function to another apparatus to continue data processing, and establishes a communication relationship between the apparatuses. Estimated value of data transfer delay that occurs during rearrangement is calculated for each of a plurality of schedulings in which the order of rearrangement of data processing functions is different, based on a storage unit that stores information indicating communication relationships. Selecting a calculation unit and a scheduling that minimizes the estimated value calculated by the calculation unit, and based on the selected scheduling, setting the order of rearrangement of the data processing functions and the start timing of the rearrangement of the data processing functions. The apparatus is characterized by comprising a setting unit and a rearrangement unit that rearranges the data processing functions to another device according to the order and timing set by the setting unit.

Further, the relocation method of the present invention is a processing device that relocates a data processing function to another device to continue data processing, and relocates the data processing function based on information indicating the communication relationship between the devices. A step of calculating an estimated value of data transfer delay occurring during rearrangement for each of a plurality of schedulings with different placement orders, selecting the scheduling with the smallest estimated value calculated, and based on the selected scheduling , setting the order of relocation of the data processing functions and the start timing of the relocation of the data processing functions, and the step of relocating the data processing functions to another device according to the set order and timing. It is characterized by

Further, the relocation program of the present invention estimates data transfer delays occurring during relocation for each of a plurality of schedulings in which the order of relocation of data processing functions is different, based on information indicating communication relationships between devices. A step of calculating a value, selecting the scheduling that minimizes the calculated estimated value, and setting the order of relocation of the data processing functions and the start timing of the relocation of the data processing functions based on the selected scheduling and causing the computer to perform the steps of relocating the data processing functions to other devices according to the established order and timing.

ADVANTAGE OF THE INVENTION According to this invention, the transfer delay which arises during rearrangement of a data processing function can be reduced.

FIG. 1 is a block diagram showing an example of the configuration of a communication system according to an embodiment. FIG. 2 is a block diagram showing an example of the configuration of the VNF operating device shown in FIG. 1. As shown in FIG. FIG. 3 is a diagram illustrating an example of a scheduling algorithm. FIG. 4 is a diagram illustrating an example of a scheduling algorithm. FIG. 5 is a diagram illustrating an example of the transit order of traffic flows to VNFs. FIG. 6 is a diagram for explaining parameters used in calculating transfer delay estimates in the parallel method. FIG. 7 shows the average added delay during VNF relocation calculated for each scheduling. FIG. 8 is a flowchart illustrating a processing procedure of VNF relocation processing according to the embodiment. FIG. 9 is a flow chart showing a processing procedure of scheduling selection processing shown in FIG. FIG. 10 is a diagram for explaining conventional rearrangement of VNFs. FIG. 11 is a diagram illustrating rearrangement of VNFs according to the embodiment. FIG. 12 is a diagram illustrating an example of a computer that implements a VNF operating device by executing a program. FIG. 13 is a diagram for explaining conventional data transfer processing. FIG. 14 is a diagram for explaining conventional data transfer processing.

Embodiments of a processing device, a rearrangement method, and a rearrangement program according to the present application will be described below in detail with reference to the drawings. Moreover, the present invention is not limited to the embodiments described below.

[Embodiment]
First, an embodiment will be described. The present embodiment relates to a relocation method for relocating virtual network functions (VNFs) implementing software activated in different servers to another device and continuing the VNFs. VNF is software that performs packet processing, such as NAPT (Network Address Port Translation), firewall, IDS (Intrusion Detection System), router, etc., and can operate on any server that has an operating environment. A VNF functions as a data processing function.

This embodiment relates to a relocation method when it is desired to operate VNFs on another server for server maintenance or load balancing. Furthermore, this embodiment describes a method of relocating VNFs to other servers when the order of traffic flow to VNFs is fixed. Note that this embodiment is not limited to rearrangement of VNFs in which the order of passage of traffic flows to VNFs is determined. It can also be applied to the migration method when

[Configuration of communication system]
FIG. 1 is a block diagram showing an example of the configuration of a communication system according to an embodiment. As shown in FIG. 1, the communication system 100 according to the embodiment includes a VNF operating device 10A and a VNF operating device 10B. The VNF operating devices 10A and 10B communicate via the network N. FIG. A host controller (not shown) is provided above the VNF operating devices 10A and 10B, and when VNFs are relocated, the VNF operating device of the relocation destination is set. Also, the number of VNF operating devices 10A and 10B is an example, and is not limited to the number shown in FIG.

The VNF operating devices 10A and 10B are physical devices that operate VNFs, specifically server devices. In this embodiment, the VNF operating device 10A will be described as a VNF relocation source device. Also, in the present embodiment, the VNF operating device 10B will be described as a VNF relocation destination device. The VNF operation device 10A is a processing device that rearranges the VNF that is being operated in its own device to the VNF operation device 10B to continue packet processing.

The VNF operating device 10A transfers the state to the relocation destination VNF operating device 10B as data for resuming the VNF in the relocation destination VNF operating device 10B. The state corresponds to NAPT flow allocation address and IDS flow behavior information, for example. The VNF operating device 10A reduces the transfer delay that occurs during VNF relocation in relation to a method of rearranging the VNFs to the VNF operating device 10B when the order of traffic flow to the VNFs is determined.

[Configuration of VNF operating device]
Next, the configuration of the VNF operating device 10A will be described. FIG. 2 is a block diagram showing an example of the configuration of the VNF operating device 10A shown in FIG. As shown in FIG. 2, the VNF operating device 10A has a communication section 11, a storage section 12 and a control section 13. FIG.

The communication unit 11 is a communication interface that transmits and receives various information to and from other devices connected via a network or the like. The communication unit 11 performs communication between another device (for example, the VNF operating device 10B) and the control unit 13 (described later) via an electric communication line such as a LAN (Local Area Network) or the Internet.

The storage unit 12 is implemented by a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, and stores a processing program for operating the VNF operating device 10A, data used during execution of the processing program, and the like. remembered. The storage unit 12 has a network parameter database (DB) 121 .

The network parameter DB 121 is a DB that manages statistical information and measurement information of events that occur within programs and networks. The network parameter DB 121 stores information indicating communication relationships between servers. The network parameter DB 121 stores, as information indicating communication relationships between servers, order information indicating the order of VNFs through which traffic flows pass.

The control unit 13 controls the entire VNF operating device 10A. The control unit 13 is, for example, an electronic circuit such as a CPU (Central Processing Unit) or an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). The control unit 13 also has an internal memory for storing programs defining various processing procedures and control data, and executes each processing using the internal memory. Further, the control unit 13 functions as various processing units by running various programs. The control unit 13 has an operation unit 131 , an estimation unit 132 (calculation unit), a scheduling unit 133 (setting unit), and a rearrangement unit 134 .

The operating unit 131 implements a packet processing function by implementing the VNF in software and operating the VNF.

The estimator 132 calculates an estimate of the transfer delay that occurs during VNF relocation. Here, as VNF scheduling, a plurality of schedulings with different VNF rearrangement orders are set. Scheduling is set using a scheduling algorithm.

The estimating unit 132 calculates an estimated value of packet transfer delay that occurs during rearrangement for each of a plurality of schedulings in which the VNF rearrangement order is different, based on the order information indicating the VNF passing order. The estimator 132 calculates an estimated value of packet transfer delay occurring during rearrangement based on the packet transfer delay between VNFs and the VNF rearrangement time. Note that the VNF relocation time is used in calculating the queuing delay.

The scheduling unit 133 selects the scheduling with the smallest estimated value among the estimated values of the packet transfer delay occurring during rearrangement calculated by the estimating unit 132 . The scheduling unit 133 sets the VNF rearrangement order and VNF rearrangement start timing based on the selected scheduling.

The rearrangement unit 134 rearranges the VNFs to the VNF operating device 10B according to the order and timing set by the scheduling unit 133 .

The VNF operating device 10B may have the same configuration as the VNF operating device 10A. The VNF operating device 10B only needs to have at least a function of operating a VNF as a VNF relocation destination.

[Example of scheduling algorithm]
An example of a scheduling algorithm will now be described. 3 and 4 are diagrams illustrating an example of a scheduling algorithm.

First, with reference to FIG. 3, a downstream priority method, which is an example of a scheduling algorithm, will be described. The downstream priority method is a method in which relocation is initiated from the VNF closer to the recipient side. The downstream side priority method is a method of sequentially rearranging each VNF with priority given to the side closer to the receiver side, that is, the downstream side. In other words, the downstream priority method, as shown in FIG. 3, when the relocation of the most downstream VNF is completed, it starts relocating the upstream VNF closest to the VNF that has completed this relocation. In this way, the downstream side priority method is a method of relocating VNFs in an order that prioritizes the downstream side.

Then, a parallel method, which is an example of a scheduling algorithm, will be described with reference to FIG. As shown in FIG. 4, the parallel method is a method of rearranging all VNFs in parallel simultaneously.

[How to calculate transfer delay estimate]
Next, a method of calculating an estimated transfer delay value performed by the estimator 132 for each scheduling algorithm will be described. First, the calculation method of the transfer delay estimate value in the downstream priority method will be described. FIG. 5 is a diagram illustrating an example of the transit order of traffic flows to VNFs. The downstream priority method is a method in which VNF3 on the downstream side has priority over VNF1 and VNF2 on the upstream side to start relocation.

In order to calculate the estimated transfer delay in the downstream priority method, an estimation formula f is set based on the average arrival interval of packets flowing through the service chain. The estimation formula f is a function for calculating the average additional delay during VNF relocation, which is the sum of the average packet queuing delay and the transfer delay through the network. The average value of packet queuing delay is represented by equation (1). A transfer delay through the network is represented by the formula (2).

where T _i is the state transfer time of VNFi. i is the VNF index number. E[I] is the average packet arrival interval. D is the transfer delay when the packet is transferred across the network between servers. In addition, T _all in equations (1) and (2) is the total state transfer time of the VNF and is expressed by equation (3).

As shown in equations (1) and (2), the estimation equation f in the downstream priority method takes T _i and E[I] as arguments. For the downstream priority method, the estimating unit 132 uses this estimation formula f to calculate the average value of the additional delay during rearrangement of transfer delays to be executed.

Next, a method of calculating an estimated transfer delay value in the parallel method will be described. When calculating the transfer delay estimate in the parallel method, an estimate equation g is set up which is a function for the calculation of the average additional delay during VNF relocation.

The estimation formula g takes T _i and E[I] as arguments. The estimation formula g is a function that indicates the average packet queuing delay. In the parallel method, the average value of packet queuing delay is given by equation (4).

T _par in the formula (4) is represented by the formula (5).

BW is a band for state transfer. S _j is the jth shortest state size. S'j _-1,j is the difference between the j-th shortest state size and the j-1th shortest state size, and is expressed by equation (6). Note that _S'0,1 = _S1 .

α(x) is a throughput correction value when state transfers of x VNFs are performed in parallel, and is expressed by the formula (7).

Here, with reference to FIG. 6, each parameter of the expressions (5) and (6) will be explained. FIG. 6 is a diagram for explaining parameters used in calculating transfer delay estimates in the parallel method.

In FIG. 6, the state transfer throughput in each section is corrected by multiplying the state transfer bandwidth BW by the corresponding α(1), α(2), and α(3). The state sizes S ₁ , S ₂ , and S ₃ are larger in the order of VNF1, VNF2, and VNF3, and the state size of each section (the section up to VNF1, the section between VNF1 and VNF2, and the section between VNF2 and VNF3) are S′ _0,1 , S′ _1,2 , and S′ _2,3 as shown in FIG. The value obtained by dividing The time required for the section up to VNF1 is _S'0,1 /(BW·α(1)), and the time required for the section between VNF1 and VNF2 is _S'1,2 /(BW·α(2 )), and the time required for the interval between VNF2 and VNF3 is S′ _2,3 /(BW·α(3)), and the state transfer time for the state with the state size S ₂ for VNF2 is the second shortest. (See (1) in FIG. 6).

Select Scheduling
Next, a method of selecting scheduling will be described. The estimator 132 calculates the average additional delay during VNF relocation corresponding to each scheduling algorithm. For example, the estimating unit 132 uses the estimation formula f in the downstream priority method to calculate the average additional delay during VNF relocation in the downstream priority method. Then, the estimating unit 132 calculates the average value of the additional delay during rearrangement of the VNFs in the parallel method using the estimation formula g in the parallel method.

Then, the scheduling unit 133 selects scheduling that minimizes the average value of additional delays during VNF rearrangement based on the calculation results of each method calculated by the estimating unit 132 .

FIG. 7 shows the average added delay during VNF relocation calculated for each scheduling. In FIG. 7, the vertical axis represents the average additional delay during VNF relocation. In FIG. 7, the horizontal axis indicates the ratio of the state size of the upstream VNF to the sum of the state sizes of the downstream VNF and the upstream VNF. Curves L1 and L2 show the average additional delay during VNF relocation calculated by the downstream priority method. Curve L3 shows the average additional delay during VNF relocation calculated by the parallel method. Curve L4 shows the average additional delay during VNF relocation calculated by the upstream-first method of relocating VNFs in upstream-first order.

A case where the state size ratio between the upstream VNF and the downstream VNF is 4:1 will be described as an example. In this case, the ratio of the state size of the upstream VNF to the sum of the state sizes of the downstream VNF and the upstream VNF is 0.8. As shown in FIG. 7, the average added delay during VNF relocation corresponding to a value of 0.8 on the horizontal axis takes the minimum value d1 in the downstream-first method shown by curve L1.

Therefore, when the state size ratio between the upstream VNFs and the downstream VNFs is 4:1, the scheduling unit 133 selects the downstream priority method. Note that, if the conditions change during relocation, the scheduling unit 133 causes the estimating unit 132 to calculate the average value of the additional delay during VNF relocation for each scheduling in accordance with the change in the conditions, and calculates the additional delay. may be changed to a scheduling that minimizes the average value of .

[VNF Relocation Processing]
Next, a processing procedure of VNF rearrangement processing by the VNF operating device 10A will be described. FIG. 8 is a flowchart illustrating a processing procedure of VNF relocation processing according to the embodiment.

When performing VNF relocation, the VNF operating device 10A transfers data necessary for VNF processing to the relocation destination VNF operating device 10B. First, as shown in FIG. 8, in the VNF operating device 10A, the rearrangement unit 134 measures the data transfer amount (state size) and transfer time associated with VNF rearrangement in the relocation destination VNF operating device 10B ( step S1). For example, based on the amount of data transferred and the measured throughput of the data transfer line, a method of estimating the transfer delay by dividing the measured value of the throughput of the data transfer line by the measured value of the transfer data amount. There is

Then, the estimating unit 132 measures the data packet transfer delay between the own device, which is the relocation source server on which the VNF to be relocated was operating, and the VNF operating device 10B, which is the relocation destination server (step S2).

Subsequently, in the VNF operating device 10A, after acquiring the data transfer time and the packet transfer delay, scheduling selection processing is performed (step S3). In step S3, the estimating unit 132 calculates an estimated value of packet transfer delay occurring during VNF relocation for each of a plurality of schedulings. Then, in step S<b>3 , the scheduling unit 133 selects the scheduling with the smallest estimated value among the transfer delay estimated values calculated by the estimating unit 132 .

In the VNF operating device 10A, the scheduling unit 133 performs scheduling processing for setting the VNF relocation order and VNF relocation start timing based on the selected scheduling (step S4). The rearrangement unit 134 rearranges the VNFs to the VNF operating device 10B according to the order and timing set by the scheduling unit 133 (step S5).

[Scheduling selection process]
Next, the processing procedure of the scheduling selection processing (step S3) shown in FIG. 8 will be described. FIG. 9 is a flow chart showing a processing procedure of scheduling selection processing shown in FIG.

As shown in FIG. 9, the estimating unit 132 receives input of the transfer delay between the relocation source VNF operating device 10A and the relocation destination VNF operating device 10B and the state size of the VNF (step S11). . Then, the estimating unit 132 calculates an estimated value of packet transfer delay that occurs during rearrangement for each scheduling (step S12). At this time, in the case of the downstream priority method, the estimating unit 132 obtains the estimation formula f using the formulas (1) to (3), and calculates the estimated value of the packet transfer delay that occurs during rearrangement. . In the case of the parallel method, the estimating unit 132 obtains an estimating formula g using formulas (4) to (7), and calculates an estimated value of the packet transfer delay that occurs during rearrangement.

Then, the scheduling unit 133 selects the rearrangement scheduling that minimizes the calculated transfer delay estimate value (step S13), and ends the scheduling selection process.

[Effects of Embodiment]
FIG. 10 is a diagram for explaining conventional rearrangement of VNFs. FIG. 11 is a diagram illustrating rearrangement of VNFs according to the embodiment.

In the conventional method, when the order of passage of the traffic flow to the packet processing function is fixed, it is not possible to avoid packet queuing delay and delay increase due to packets passing through the network during VNF relocation, As shown in FIG. 10, packet transfer delays occur during VNF relocation. For this reason, the conventional method cannot smoothly reflect the user's actions in the virtual space for a VR or AR user who has moved a long distance.

On the other hand, the VNF operating device 10A according to the present embodiment performs rearrangement for each of a plurality of schedulings with different VNF rearrangement orders based on order information indicating the passage order of the VNFs through which the traffic flow passes. Compute an estimate of the intervening packet transfer delay. Then, the VNF operating device 10A sets the VNF rearrangement order and VNF rearrangement start timing based on the scheduling that minimizes the calculated transfer delay estimate value, and according to the set order and timing, Relocate the VNF to the VNF operating device 10B.

In this way, the VNF operating device 10A reduces the transfer delay when a packet is transferred between servers and the queuing delay that occurs when a packet arrives at a VNF being relocated by changing the scheduling for VNF relocation. can be improved. Therefore, as shown in FIG. 11, the VNF operating device 10A can suppress the delay incurred by packets during relocation of VNFs more than the conventional method, and can maintain the quality of experience of services that require low-delay communication. can.

As a result, the VNF operating device 10A can reduce the transfer delay during relocation of servers and VNFs on paths for server load balancing and maintenance. Even when providing AR or VR, which requires data from the processing server to be transmitted to the terminal with a short delay in order to be experienced in real life, the quality of experience is not impaired.

Therefore, according to the VNF operating device 10A, it is possible to reduce the transfer delay that occurs during rearrangement of the packet processing functions when the traffic flow passing order to the packet processing functions is determined.

The present embodiment is not limited to the relocation of VNFs in which the order of passage of traffic flows to VNFs is determined, but is also applicable when a virtual machine having a data processing function for processing data such as an application layer migrates to another server device. It is possible to reduce the transfer delay that occurs during rearrangement of data processing functions.

[About the system configuration of the embodiment]
Each component of the VNF operating devices 10A and 10B shown in FIG. 1 is functionally conceptual, and does not necessarily need to be physically configured as shown. That is, the specific forms of distribution and integration of the functions of the VNF operating devices 10A and 10B are not limited to those shown in the drawings, and all or part of them can be functionally distributed in arbitrary units according to various loads and usage conditions. Alternatively, it can be physically distributed or integrated.

Further, each process performed in the VNF operating devices 10A and 10B may be implemented entirely or in part by a CPU and a program analyzed and executed by the CPU. Further, each process performed in the VNF operating devices 10A and 10B may be implemented as hardware based on wired logic.

Moreover, among the processes described in the embodiments, all or part of the processes described as being automatically performed can also be performed manually. Alternatively, all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the above-described and illustrated processing procedures, control procedures, specific names, and information including various data and parameters can be changed as appropriate unless otherwise specified.

[program]
FIG. 12 is a diagram showing an example of a computer that implements the VNF operating devices 10A and 10B by executing a program. The computer 1000 has a memory 1010 and a CPU 1020, for example. Computer 1000 also has hard disk drive interface 1030 , disk drive interface 1040 , serial port interface 1050 , video adapter 1060 and network interface 1070 . These units are connected by a bus 1080 .

Memory 1010 includes ROM 1011 and RAM 1012 . The ROM 1011 stores a boot program such as BIOS (Basic Input Output System). Hard disk drive interface 1030 is connected to hard disk drive 1090 . A disk drive interface 1040 is connected to the disk drive 1100 . A removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1100 . Serial port interface 1050 is connected to mouse 1110 and keyboard 1120, for example. Video adapter 1060 is connected to display 1130, for example.

The hard disk drive 1090 stores an OS (Operating System) 1091, application programs 1092, program modules 1093, and program data 1094, for example. That is, a program that defines each process of the VNF operating devices 10A and 10B is implemented as a program module 1093 in which code executable by the computer 1000 is described. Program modules 1093 are stored, for example, on hard disk drive 1090 . For example, the hard disk drive 1090 stores a program module 1093 for executing processing similar to the functional configuration in the VNF operating devices 10A and 10B. The hard disk drive 1090 may be replaced by an SSD (Solid State Drive).

Also, the setting data used in the processing of the above-described embodiment is stored as program data 1094 in the memory 1010 or the hard disk drive 1090, for example. Then, the CPU 1020 reads out the program module 1093 and the program data 1094 stored in the memory 1010 and the hard disk drive 1090 to the RAM 1012 as necessary and executes them.

The program modules 1093 and program data 1094 are not limited to being stored in the hard disk drive 1090, but may be stored in a removable storage medium, for example, and read by the CPU 1020 via the disk drive 1100 or the like. Alternatively, program modules 1093 and program data 1094 may be stored in another computer connected via a network (LAN (Local Area Network), WAN (Wide Area Network), etc.). Program modules 1093 and program data 1094 may then be read by CPU 1020 through network interface 1070 from other computers.

Although the embodiments to which the invention made by the present inventor is applied have been described above, the present invention is not limited by the descriptions and drawings forming part of the disclosure of the present invention according to the embodiments. That is, other embodiments, examples, operation techniques, etc. made by those skilled in the art based on the present embodiment are all included in the scope of the present invention.

100 communication system 10A, 10B VNF (Virtual Network Function) operating device 11 communication unit 12 storage unit 13 control unit 121 network parameter database (DB)
131 Operation Unit 132 Estimation Unit 133 Scheduling Unit 134 Rearrangement Unit

Claims (6)

A processing device that relocates data processing functions to another device to continue data processing,
a storage unit that stores information indicating communication relationships between devices;
a calculation unit that calculates, based on the information indicating the communication relationship, an estimated value of a data transfer delay occurring during rearrangement for each of a plurality of schedulings in which the order of rearrangement of the data processing functions is different;
Selecting the scheduling that minimizes the estimated value calculated by the calculating unit, and setting the order of rearrangement of the data processing functions and the start timing of the rearrangement of the data processing functions based on the selected scheduling. a setting unit for
a relocation unit that relocates the data processing functions to the other device according to the order and timing set by the setting unit;
A processing apparatus comprising: 2. The processing apparatus according to claim 1, wherein said storage unit stores, as information indicating a communication relationship between said apparatuses, order information indicating a passage order of said data processing functions through which a traffic flow passes. The calculation unit calculates an estimated value of the data transfer delay occurring during the relocation based on the data transfer delay between the data processing functions and the relocation time of the data processing functions. 3. The processing apparatus according to claim 1 or 2. 1-, wherein the plurality of scheduling includes a method of starting rearrangement from the data processing function closer to the receiver side, and a method of starting rearrangement simultaneously for all data processing functions. 4. The processing apparatus according to any one of 3. A processing device that relocates data processing functions to another device to continue data processing,
calculating an estimated value of data transfer delay occurring during rearrangement for each of a plurality of schedulings in which the order of rearrangement of the data processing functions is different, based on information indicating communication relationships between devices;
selecting the scheduling that minimizes the calculated estimated value, and setting the order of rearrangement of the data processing functions and the start timing of the rearrangement of the data processing functions based on the selected scheduling; ,
relocating the data processing functions to the other device according to a set order and timing;
A relocation method characterized by including calculating an estimated value of data transfer delay occurring during rearrangement for each of a plurality of schedulings in which the order of rearrangement of data processing functions is different, based on information indicating communication relationships between devices;
selecting the scheduling that minimizes the calculated estimated value, and setting the order of rearrangement of the data processing functions and the start timing of the rearrangement of the data processing functions based on the selected scheduling; ,
relocating the data processing functions to other devices according to a set order and timing;
A relocation program that causes a computer to run

Images